> Source: https://tryordinary.com/blog/ab-test-profit-and-ad-spend-for-shopify-brands A/B testing [All articles↗](/blog) # See which test version made you more money, and what the traffic cost August 21, 2026 ## Overview Conversion rate is where most A/B tests stop. But the version that converts best isn't always the one that makes you the most money. Say your test variant adds a free gift: more shoppers check out, and each order costs you more to fulfill. So we took our [experiment results](/blog/ab-testing-for-shopify-brands) further. Every test now reads out in profit, sits next to the ad spend that paid for its traffic, and watches the metrics a winning variant can quietly damage. You pick the version that grows the business, not just the one with the prettiest conversion rate. What you see ## A test result you can take to the bank ### Profit for each version Each version's revenue is followed all the way down to profit, using the product costs you've already entered: what the goods cost, shipping, and payment and fulfillment fees, ending in the contribution profit each version earned. A version that converts better on a deep discount can still make you less money, and this is where you see it. ### The ad spend behind your test See what you spent on Meta and Google during the test, next to the new customers each version brought in and the cost per new customer for each. Test results and ad spend sit on one page, so you can tell whether a winning version also makes your paid traffic work harder. ### Guardrails that catch the hidden cost Refund rate, add-to-cart rate, bounce rate and 30-day repeat purchase are tracked for each version, with a clear flag when one of them gets worse. A variant that lifts orders but brings more returns doesn't get to hide behind its headline number. ### The funnel, step by step Each version's path from session to add to cart, to checkout, to order is laid out side by side, with the rate at every step. You can see exactly where one version pulls ahead or falls behind, and a clear verdict badge tells you where the test stands. Also in this release ## Compare any date range Compare any date range with the same period last year, or place a custom comparison window anywhere on the calendar. Comparisons line up by weekday, so Mondays sit against Mondays and your weekend mix stays fair, and they carry through your reports and scorecards. Available on every plan. ## Available now The experiment results page is on **every plan**; the Free plan includes one A/B test. Ad spend context appears on the **Starter plan and above** with Meta or Google Ads connected. Profit per version is on the **Advanced plan** once your product costs are in. [Connect your store](https://apps.shopify.com/ordinary-attribution-ab-testing) [Compare plans](/pricing) Keep reading ## More from the blog [October 16, 2026 Stop — it's losing: a clear call on A/B tests that won't recover Ordinary's A/B test results now tell Shopify brands when to stop a losing version, with the chance it still wins, and name the leader of overlapping tests.](/blog/stop-a-losing-test)[October 9, 2026 Landing-page angles: every version of a page, with what it earns See each landing-page angle and split-URL test arm as a row in Page Performance, with visits, orders, conversion rate, spend, CAC and ROAS per version.](/blog/landing-page-angles)[October 2, 2026 Reach rate: score homepage tests on where visitors go next Score homepage and landing-page A/B tests on how many visitors reach your product pages, and run several tests at once without them interfering.](/blog/ab-testing-reach-rate) ## Pick the version that pays. Connect your Shopify store and read your next A/B test in profit, not just conversion rate. [Connect your store](https://apps.shopify.com/ordinary-attribution-ab-testing)